Cezar D.

@dalmaer Turns out each is best suited for their respective species x.com/i/grok/share/0…

Funny how “God” worked it out for the octopus but not humans. Huh.

@Dimillian Barely. Only because I have no easy way to do that.

Let’s be honest here, how many time this week have you directly @ a code file to your agent?

@ryanvogel @jayair how many days since the latest Vogel incident?

updated setup

@thekitze What else?

using gpt 5.4 for openclaw

@kentcdodds Did you build the internet???

I built this by myself 😅

@thdxr *this* close to a dbrand ad

what about armor but for your laptop is that anything?

@j1ngb0 @thdxr @krishstwt @opencode what you trippin on bro

@thdxr @krishstwt @opencode ain’t nobody coding from ios lol. just let me tag opencode in slack and push straight to pr pls

. @thdxr by any chance will there be an ios app for @opencode ?

@thdxr What's a devrel?

ok you know what im gonna ruin this bit for you all and stop fighting it im the best devrel to ever devrel im king devrel, no one has ever done it like me your mom? already had developer relations with her come get it

@trq212 This plugin has existed since last year: github.com/backnotprop/pl…

New in Claude Code: /ultraplan Claude builds an implementation plan for you on the web. You can read it and edit it, then run the plan on the web or back in your terminal. Available now in preview for all users with CC on the web enabled.

@trq212 @thdxr So you basically implemented a feature that's existed as an opencode plugin for half a year probably? github.com/backnotprop/pl…

@thdxr no but let me know if you have any questions about ultraplan 👨‍💼

This is my favorite one

@thekitze Yep, nobody harnesses gpt-5.4 properly except codex

codex works better in the codex cli/app than in cursor and you cannot convince me otherwise. i dont remember the last time i yelled at an llm

@thekitze @peduarte Let me guess, you found 2 weeks ago about it

@peduarte I singlehandedly keep the hype alive

actual video of @thekitze being part of the rayfam

@thekitze Do a poll bro

are u using codex 5.3 or gpt 5.4 for coding

@thekitze @raycast What do you use it for?

TIL @raycast supports local models 🤯 it has native ollama integration but it also supports lm studio by enabling "custom providers" and adding the openai compatible endpoint to the yaml file 🤓 dope!!!

@thsottiaux @badlogicgames I would definitely buy that. Even a 50$ one

@badlogicgames 👀

yo, openai, now is the time to introduce a $100/month codex subscription.

@AaronHenray @thsottiaux I concur

Is something broken with the counting of token usage? I'm down to 15% left on the Pro plan in 3 days working on tweaking one app project. I've struggled in the past to use 50% of Pro limits in a week while working on multiple large from scratch builds. Usually on Plus that lasts me fine for about 4-5 days if I am using it the way I am now. 🧐

From the surge in demand on Codex and many other services, it is clear that AI companies are going to go through a phase of demand outpacing supply. Two things will be most important, having the most capacity and having the most efficient models. I believe we have both.

@thsottiaux Seems 5h quota is draining much faster than 2x, more like 4x tbh. Not sure why, but if it's because of peak-hours, it should say what's the "schedule". BTW, I love Codex, but not sure I'm able to go as high as $200... $50 and $100 plans would be nice...

With Codex the there is quite the gulf in load between peak and off-peak times, and we would like to achieve more of a smoother traffic pattern as that would be a more optimal use of our compute. We have ideas, but curious what you all think we should do? Would more usage during off-peak and surge multiplier during peak times make sense?

@McKamyk @vimtor Sst * (not set 😅)

@McKamyk @vimtor AFAIK they base opencode (and probably set) on opentui, also maintained by them

we shipped several tui improvements: - filter logs to specific lambda functions - shortcuts to clear and scroll to the bottom - drag for easier copy-paste you should also see less flickering and better errors